Evaluations Aggregate to equal UI

Hello,

I am currently reviewing the evaluations aggregate API against the PureCloud UI. The Aggregate API without filters includes evaluations that were not released (set to NeverRelease= true). Is there a way to filter to remove those from the total as the PureCloud UI does?

POST /api/v2/analytics/evaluations/aggregates/query

Example Body
{
"interval": "2020-01-01T00:00:00.000Z/2020-02-01T00:00:00.000Z",
"groupBy": [],
"views": []
}

Per the docs for POST /api/v2/analytics/evaluations/aggregates/query, the valid values for a filter predicate dimension are:

calibrationId, contextId, conversationId, divisionId, evaluationCreatedDate, evaluationId, evaluatorId, formId, queueId, rescind, rescored, userId

You can request the ability to filter by additional metrics and share your use case at https://purecloud.ideas.aha.io/ideas.

Thanks Tim,

As a follow up do you happen to know how the totals in the Genesys Cloud UI are collected from the API as we are attempting to match them? As discussed we cannot use the aggregate and pulling the detail would not be the most efficient way.

It depends on where you're looking. Reports and views are documented on the resource center: https://help.mypurecloud.com/articles/about-reports-views-and-dashboards/

Hi Tim,

What I'm looking for is something more low level that tells me the API used so that my customer can replicate the numbers. A bit like the devcon session on how reports are generated using the API.

The reports and views document how each column or field is calculated, though it's in written English instead of API citations. If you need clarification on a particular data point, let me know which one.

Hi Tim,

I have no issue with the calculation except the Performance View (not looking at the report) only counts released evaluations where the aggregate API includes ones not released or set to never release. I'm trying to see exactly how the Performance View uses the API for that as I can only assume it is using the detail API which would seem far more work.

I don't have evaluations configured in my org to verify the counts, but I believe the evaluations count in the Agent Evaluation Summary View is coming from GET /api/v2/quality/agents/activity.

When the view documentation isn't specific enough, check the network tab in the dev tools in your browser and you can see the exact requests made by the UI as you load various views.

That's what I came up with. It took me a minute to find it as the process is different to the other queries. I assume this was prior to the aggregates API being available, We can use that for now. I have an idea accepted to update the Evaluations View which will likely mean a new API or updates to the existing.

This topic was automatically closed 31 days after the last reply. New replies are no longer allowed.